Review of remaining useful life prediction using support vector machine for engineering assets

Remaining useful life (RUL) is important to manage life circles of machineries and reduce maintenance cost. Support vector machine (SVM) is a promising algorithm for RUL prediction because of its advantages to deal with small size of training sets and multi-dimensional data. Recently, many methods of RUL prediction using SVM have been proposed. In this paper, a review over 60 references within the last 10 years on this topic is conducted, which includes introduction of the improvement algorithms and applications of using SVM to predict RUL and possible problems to be solved in future work.